The prefrontal cortex (RFC) is critical for adaptive higher-order cognitive behaviors that are compromised by a wide variety of mental health disorders including schizophrenia, (ADHD), substance abuse disorders, Alzheimer's and Parkinson's Disease, and AIDS-related dementia. A better understanding of basic neural mechanisms will lead to improved diagnostic, prognostic, and therapeutic procedures. Although the RFCis critical for the planning, maintenance, selection, and execution of willed behavior, we know very little about the mechanisms by which it accomplishes these goals. Barriers to our progress in this regard include 1) a poor understanding of how the crucial animal work on RFC functions translates to the human species we are trying to understand, and 2) a lack of understanding of how the RFC influences ongoing behavior through its functional interactions with other brain areas. Here we propose a divide-and-conquer strategy for better understanding the functions of the RFC. In AIM 1, we will localize a key portion of the RFC, the human homolog of the monkey frontal eye field (FEF) and treat it as a model system for detailed study of RFC functions. We strategically chose the FEF as our model because 1) unlike other RFC areas, we have methods for localizing it in humans, 2) data from monkey FEF, as compared to other RFC areas, offer testable predictions about the functional homologies between the species, and most importantly 3) the FEF is implicated in many of the same high-level cognitive behaviors that the RFC in general is implicated. We will study the mechanisms that the human FEF uses for planning, attention, memory, and selection. Working within a better-defined and constrained system like the oculomotor system may quickly lead to mechanistic accounts of these functions that may be less tenable in a more complicated and less understood system like the RFC as a whole. Although the RFC is thought to influence ongoing behavior through its functional interactions with other brain areas, there is a dearth of evidence to support this theory. In AIM 2, we will use fMRI to measure functional interactions between the RFC and other brain areas that together may form networks supporting the critical behaviors. Together, the two AIMS embrace both functional specialization at the local level and distributed processing at the network level and will allow us to test critical hypotheses about how the RFC supports intention, attention, and working memory.